VADER: Visually Aimed Defeasible Reasoning

overview

The project currently contains an official implementation of the article "VADER; Visually Aimed Defeasible Reasoning".

Quickstart

TBD how to use the evaluation_all.py script with a few examples

experiments

according to a few executions, the prompt

"Question: The following is a multiple choice question. Choose an answer by it's number
     ...: \n 1.There is a tower in the image\n 2. There is a castle in the image.\n Answer:"

returns always A.

https://discuss.huggingface.co/t/trying-to-understand-system-prompts-with-llama-2-and-transformers-interface/59016

further notes issues and FAQs

For installation and running of LLaMA2 using transformers python package (by Huggingface.co) you will need to create a read token for the LLaMA resource. More information can be found here

blogpost